| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
well, just doing now what we can
|
| |
|
| |
|
|
|
|
| |
separate so you can for example have -harddisk1 as ID 6 and -harddisk 2 as ID 5. The HLE'd CD & HD have configuration switches to set the ID, real emulated hardware will specify the ID using it's own method. [smf]
|
|
|
| |
stuff into bebox and model3 classes. nw.
|
| |
|
| |
|
|
|
|
| |
everything to act as a PC. A copy of the VIA VT83C461 datasheet would be useful to remove some of the hacks added to make the games using it work again. [smf]
|
| |
|
| |
|
|
|
|
| |
--
The apollo and bebox both boot but might need more thorough testing.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
also use am9517a device now (no whatsnew)
|
|
|
|
|
|
| |
(MESS) bw2: Fixed memory mapping. (nw)
(MESS) msm6255: Refactored to use device_memory_interface. (nw)
(MESS) floppy: Added macros for declaring modern floppy formats. The generic floppy formats (D88/DFI/IMD/IPF/MFI/MFM) are now automatically supported in each driver using the modern floppy code. [Curt Coder]
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Remaining TODO list:
- take WP into account
- test the amstrad, implement its observational format (edsk) using
pasti as a start. Or find the legendary amstrad IPFs. Or both.
- correct read track, the implementation is completely wrong. See
previous for testing, it's only used in protections the check the
inter-sector gaps.
- shake and bake on the amstrad, protections are the best to find bugs
in a fdc
- add the scan id commands, but nothing seems to use them
- debug the 2.88M formatting which is unreliable. Fix its IDAM/DAM
gap size on formatting too (but that's not what's making it
unreliable)
- test all the systems that were hit, and fix what needs to be fixed.
Beware that multiple problems may happen:
- upd765 may be wrong
- the driver may not be working
- the hookup may be wrong/incomplete (bitrate selection and floppy
rpm in particular)
- the driver may be too limited for the new implementation (the x68k
dma device does not handle non-instant dma yet for instance)
- report invalid command when appropriate depending on the actual chip
emulated
- add the russian clones with their real names
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
to 8-bit handlers to match the others. To ease
pain, added DECLARE_READ/WRITE_HANDLER macros that
set up a default parameter. Also updated devcb so
that the handlers can be called with or without the
mem_mask. [Aaron Giles]
|
|
|
| |
note: hoarded dump removed too from coco_cart.xml, this will not be tolerated
|
| |
|
|
|
|
| |
changed to be members of state classes (no whatsnew)
|
|
|
|
| |
MESS. (nw)
|
| |
|
|
|